ACTA UNIVERSITATIS
UPSALIENSIS
Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Medicine
1754
Sleep disturbances
Consequences and comorbidities
SHADI AMID HÄGG
ISSN 1651-6206 ISBN 978-91-513-1244-6
Dissertation presented at Uppsala University to be publicly examined in H:son
Holmdahlsalen, 100/101, 2 tr, Uppsala, Thursday, 7 October 2021 at 09:00 for the degree of Doctor of Philosophy. The examination will be conducted in Swedish. Faculty examiner: Professor Göran Kecklund (Stockholms Universitet, Stressforskningsinstitutet vid
psykologiska institutet).
Abstract
Amid Hägg, S. 2021. Sleep disturbances. Consequences and comorbidities. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1754. 81 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-1244-6.
Background: Sleep disorders are common in the general population, with insomnia and
sleep-related breathing disorders being the most common disorders. Since sleep has many important functions, such as a role in consolidation of memories and learning, energy conservation, cardiovascular and immune system regulation, it is not surprising that the disruption of normal sleep may lead to negative health effects and various comorbidities.
Aim: The overall aim of this thesis was to investigate the impact of disturbed sleep on various
consequences and comorbidities.
Methods and results: Papers I and II were based on the Sleep and Health in Women (SHE),
a population-based prospective study of women, where a questionnaire was sent to women in 2000 and 2010.
In paper I, the study cohort comprised 4,320 women <67 years of age who answered both questionnaires and had worked during the follow-up period. In women, having a long history of insomnia symptoms was associated with an increased risk of self-reported occupational accidents.
In paper II, the 4,882 participants who answered the questions regarding nocturnal gastroesophageal reflux and snoring in both questionnaires were included in the study cohort. Women with nocturnal gastroesophageal reflux and snoring were at an increased risk of developing daytime sleepiness and to involuntarily fall asleep during the day.
Paper III was based on the RHINE-cohort with participants from seven Northern European centers. The study cohort in paper III comprised the 2,568 smokers in the baseline study that also reported being smokers or former smokers in the follow-up study. It was found that having insomnia symptoms or excessive daytime sleepiness decreases the chance of long-term smoking cessation, and that smoking increases the risk of incident difficulties inducing sleep.
Paper IV was the population-based, cross-sectional GA2LEN-survey which was conducted in four major Swedish cities. Paper IV included the 25,901 participants who answered questions regarding both snoring and insomnia symptoms. The combination of snoring and insomnia symptoms was associated with an increased risk of hypertension, asthma, chronic obstructive pulmonary disease, and daytime sleepiness.
Conclusions: Disturbed sleep, due to varying causes, influences the risk of occupational
accidents, on the chance of successful smoking cessation, on the risk of daytime sleepiness, hypertension, and obstructive lung disease. In clinical consultation, it is important to always inquire about disturbed sleep as it can have an impact on many aspects of health.
Keywords: Insomnia, sleep, reflux, snoring, occupational accidents, smoking, hypertension Shadi Amid Hägg, Department of Medical Sciences, Lung- allergy- and sleep research, Akademiska sjukhuset, Uppsala University, SE-75185 Uppsala, Sweden.
© Shadi Amid Hägg 2021 ISSN 1651-6206 ISBN 978-91-513-1244-6
There are some nights when
sleep plays coy,
aloof and disdainful.
And all the wiles
that I employ to win
its service to my side
are useless as wounded pride,
and much more painful.
Maya Angelou, Insomniac
List of Papers
This thesis is based on the following papers, which are referred to in the text by their Roman numerals.
I Hägg SA, Torén K, Lindberg E. Role of sleep disturbances in occupational accidents among women. Scand J Work Environ Health, 2015;41(4):368-76
II Hägg SA, Emilsson OI, Franklin K, Janson C, Lindberg E. Noc-turnal gastroesophageal reflux increases the risk of daytime sleepiness in women. Sleep Med. 2019;53:94-100
III Hägg SA, Ljunggren M, Janson C, Holm M, Franklin KA, Gisla-son T, Johannessen A, Jõgi R, Olin AC, Schlünssen V, Lindberg E. Smokers with insomnia symptoms are less likely to stop smok-ing. Respiratory Med. 2020;170:106069
IV Hägg SA, Ilieva E, Ljunggren M, Franklin KA, Middelveld R, Lundbäck B, Janson C, Lindberg E. The negative health effects of a combination of snoring and insomnia (submitted)
Contents
Introduction ... 11
Normal sleep ... 11
Sleep stages and architecture ... 11
Sleep regulation ... 13
Sleep disorders ... 14
Insomnia ... 15
Sleep-related breathing disorders ... 18
Diagnostic methods ... 20
Objective assessment of sleep and daytime symptoms... 20
Subjective assessment of sleep and daytime sleepiness ... 21
Consequences and co-morbidities ... 23
Daytime sleepiness ... 23
Accidents ... 25
Gastroesophageal reflux and sleep ... 30
Cardiovascular diseases and sleep ... 32
Cigarette smoking and sleep ... 35
Aims ... 36
Methods ... 37
Study populations ... 37
SHE (Papers I & II) ... 37
RHINE (Paper III) ... 38
GA2LEN (Paper IV) ... 39
Questionnaires and definitions ... 39
Overview of questionnaires ... 39
Insomnia symptoms and snoring ... 39
Daytime sleepiness ... 40 Smoking ... 41 Covariates ... 42 Register data ... 42 Study designs ... 42 Statistical methods ... 44 Ethics ... 44 Results ... 45 Paper I ... 45
Paper II ... 46
Paper III ... 48
Paper IV ... 48
Discussion ... 51
Methodological considerations ... 54
Clinical implications and future perspectives ... 56
Conclusions ... 59
Summary in Swedish - Sammanfattning på svenska ... 60
Acknowledgements ... 63
Abbreviations
BNSQ Basic Nordic Sleep Questionnaire
BMI Body mass index
CAGE Cut down, Annoyed, Guilty, Eye-opener
COPD Chronic obstructive pulmonary disease
DIS Difficulty initiating sleep
DMS Difficulty maintaining sleep
DSM Diagnostic and Statistical Manual of Mental disorders
ECRHS European Community Respiratory Health Survey
EDS Excessive daytime sleepiness
EMA Early morning awakening
ECG Electrocardiogram
EEG Electroencephalogram
ESS Epworth Sleepiness Scale
GABA Gamma-aminobutyric acid
GA2LEN Global allergy and asthma network of excellence
GERD Gastroesophageal reflux disease
ICD International Classification of Sleep Disorders
IFA Involuntary falling asleep
ISI Insomnia Severity Index
MCH Melanin-concentrating hormone
MnPN Median preoptic area
MSLT Multiple sleep latency test
MWT Maintenance of wakefulness test
nGER Nocturnal gastroesophageal reflux
NREM Non-rapid eye movement
OSA Obstructive sleep apnea
PSG Polysomnography
PSQI Pittsburg Sleep Quality Index
RAS Reticular activating system
REM Rapid eye movement
RHINE Respiratory Health In Northern Europe
SDB Sleep-disordered breathing
SCN Superchiasmatic nucleus
SHE Sleep and HEalth in Women
VLPO Ventrolateral preoptic area
UARS Upper airway resistance syndrome
Introduction
As far back as in ancient Greece, mankind has had an interest in and awareness of the importance of sleep. In Greek mythology Hypnos, the god of sleep, is the son of Nyx, the god of darkness and Erebus, the night goddess. His twin-brother was Thanathos, the god of death, indicating the close relationship the ancient Greeks believed there was between sleep and death.1, 2 In contrast to the earlier Greeks’ divine explanations of sleep, the “father of medicine” Hip-pocrates and his disciples considered sleep as an important part of prognosis and diagnosis and important for the maintenance of health.1
“As for sleep, the patient ought to follow the natural custom of being awake during the day and asleep during the night. Should this be changed it is rather a bad sign…The worst thing is not to sleep either during the day or during the night. For either it will be pain and dis-tress that cause the sleeplessness or delirium will follow this symp-tom”
Hippocrates. Prognostic. 500 BC.3
In spite of the early Hippocratic theories regarding sleep and health, it was not until the 20th century that sleep research took off, a milestone being the first recording of electroencephalogram (EEG) wave patterns from the brain in 1924.4 The German psychiatrist Hans Berger used the EEG to study the dif-ference between sleep and wakefulness.4 During the 1950s rapid eye move-ment (REM-sleep) was discovered and more sleep research was performed. However, it was not until the 1970s that sleep medicine was designated a clin-ical medclin-ical field.5
Normal sleep
Sleep stages and architecture
For adults between the ages of 18 and 60 years the recommended amount of sleep is at least 7 hours per night, but about one-third of adults sleep less than 7 hours.6-8 Normal sleep consists of two different types of sleep; the non-rapid eye movement (NREM) and the rapid eye movement (REM) sleep. NREM is further subdivided into three stages according to the depth of sleep: N1, N2 and N3. N1 is the transition from wakefulness (W) to light sleep (N2) and N3
is deep sleep (Figure 1). The differentiation between the stages is based on specific brainwave activity and neuronal activity.
REM sleep differs from NREM by the occurrence of atonia in skeletal mus-cles, burst of rapid eye movement and by a waking-like EEG pattern.9 During a normal night sleep oscillates between NREM and REM sleep approximately every 60–90 minutes.9
Usually, in adults, the first cycle of the night starts with a transition into stage N1, N2, N3 and then REM. The percentage of REM sleep in each cycle in-creases as the night progresses while the amount of deep sleep dein-creases.10 Arousals are short interruptions of sleep lasting 3–15 seconds and normally occur around 10–20 times/hour throughout the night. An interruption of sleep that lasts more than 30 seconds is considered an awakening.10
From childhood to adulthood, the total sleep time and the proportion of deep sleep and REM-sleep decreases.11
Stage
Hours of sleep
1 2 3 4 5 6 7 8
Figure 1. Schematic depiction of normal hypnogram in adult.
Wakefulness REM N1 (transition) N2 (light sleep) N3 (deep sleep) NREM Arousals
Sleep regulation
The regulation of sleep and wakefulness is complicated; even though it is still not fully understood, great progress has been made during the last decades in understanding its pathways.
A currently accepted model for sleep regulation proposes an interaction or opposition between two regulatory processes; process S (the homeostatic pro-cess/sleep pressure) and process C (the circadian pacemaker).12 The longer the duration of wakefulness, the greater the pressure to sleep.12
Adenosine accumulation during wakefulness is thought to play an important role in increasing sleep pressure.13
The circadian pacemaker, located in the suprachiasmatic nucleus (SCN) of hypothalamus, fluctuates with a cycle of about 24 hours independent of the amount of preceding sleep or wakefulness13. The SCN is affected by exoge-nous and endogeexoge-nous factors (zeitgebers), the most important one being ocular light exposure.14 The SCN receives information regarding daylight from the retina through the retinohypothalmatic pathway and the information is then transmitted to centers in the brain that maintain sleep or waking.13 When there is a lack of daylight stimuli, SCN acts on the pineal gland to increase the se-cretion of melatonin. Melatonin, in turn, acts reciprocally on the SCN to in-hibit its wake-promoting effect, thus promoting sleep.15
The main sleep-promoting areas in the brain, the ventrolateral preoptic area (VLPO) and median preoptic area (MnPN), are located in the hypothalamus and have both efferent and afferent connections to the arousal systems.16 The activation of the sleep-promoting areas leads to the release of inhibitory neu-rotransmitters galanin and gamma-aminobutyric acid (GABA) that promote sleep by inhibiting the arousal systems in the brainstem and basal forebrain.16 The body’s main arousal system, the reticular activating system (RAS) consist of a neural network in the brainstem and has efferent pathways to the basal forebrain, hypothalamus and cortex.16 It works through two different arousal pathways; the monoaminergic arousal system and the cholinergic arousal sys-tem. The most important neurons in the monoaminergic arousal system are the dopaminergic nuclei (periaqueductal grey), noradrenergic nuclei (locus co-eruleus), serotonergic nuclei (raphe nuclei) and histaminergic nucleus (tuber-omammillary) .13 The cholinergic system consists of the laterodorsal tegmen-tal nucleus and the pedunculopontine nucleus.13 The neurons in these two systems are highly active during wakefulness and have reduced activity during sleep.17 Another wake-promoting part of the brain consists of anticholinergic neurons in the basal forebrain.17
Orexins are peptides expressed in neurons within the lateral hypothalamus and their proposed function is to stabilize wakefulness.17 Efferent pathways from orexin neurons to the monoaminergic arousal system promote wakefulness. Afferent pathways from the VLPO inhibit the release of orexin, and thereby promote sleep.17 As a counterpart to orexins, neurons in the hypothalamus produce melanin-concentrating hormone (MCH) which acts as a sleep-pro-moter through an inhibitory effect on the nuclei in RAS.18 Figure 2 schemati-cally depicts the pathways of sleep and wakefulness.
The interaction between the monoaminergic nuclei that promote wakefulness and the sleep promoting nuclei in VLPO is proposed to act as a “flip-flop-switch” for the transitions between wakefulness and sleep and orexin is de-scribed as the stabilizer of the switch.13, 19
Sleep disorders
There are several nosology systems that classify sleep disorders into various diagnosis. The third edition of the International Classification of Sleep
Disor-BASAL FORE-MIDBRAIN PONS MEDULLA BRAINSTEM HYPOTHALAMUS Orexins neurons MCH neurons Preoptic area PINEAL GLAND Melatonin
+
-
GABA, galanin-
-
-Orexin+
MCH-
-
Figure 2. Schematic of the pathways involved in the regulation of sleep and wake-fulness. [MHC = melatonin concentrating hormone, RAS= reticular activating
ders (ICSD-3, 2014) and the Diagnostic and Statistical Manual of Mental dis-orders (DSM-5, 2013)20, 21 are the two major ones. In addition, for coding of sleep disorders the International Classification of Diseases (ICD-11, 2020) may be used.22
ICSD-3 classifies sleep disorders into seven main groups while DSM-5 con-siders the underlying pathological and etiological factors and has ten main diagnostic groups20, 21 (Table 1). For example, in ICSD-3, parasomnias are main group, which is then further subdivided into several diagnosis. In DSM-5, the umbrella concept of parasomnias is not used; instead the main diagnosis groups of NREM-sleep arousal disorder, REM-sleep behavior disorder and nightmare disorder are used.
Table 1. The major categories of sleep disorders according to DSM-5 and ICSD-3
DSM-5 ICSD-3
Insomnia disorder
Breathing-related sleep disorders Hypersomnolence disorder Narcolepsy
Circadian rhythm sleep wake disorders NREM sleep arousal disorders REM sleep behavior disorder Nightmare disorder
Restless legs syndrome
Substance/medications-induced sleep disor-der
Insomnia
Sleep–related breathing disorders Central disorders of hypersomnolence Circadian rhythm sleep-wake disorders Parasomnias
Sleep-related movement disorders Other sleep disorders
Sleep disorders are common in the general population and studies from coun-tries across the globe show the prevalence to be between 3.9–56%.23, 24 Of the sleep disorders, insomnia and the sleep-related breathing disorders (mainly obstructive sleep apnea) are the most common. Thus the focus will be on these disorders in the remainder of this thesis.
Insomnia
Definition
In both DSM-5 and ICSD-3, a diagnosis of insomnia requires a report of sleep initiation or sleep maintenance problems, despite adequate opportunity to sleep and with daytime consequences as the result of the insomnia symp-toms20, 21 (Table 2). In ICSD-3, the insomnia disorder is considered chronic if it lasts more than three months and short-term if it lasts for less than three months. The corresponding terminology in DSM-5 is persistent and transient
insomnia. The most important change between the previous editions of both ICSD and DSM (ICSD-2 and DSM-4) is that the concept of primary and sec-ondary disorders have been removed from the nosology system.20, 21, 25, 26 There were several reasons for this paradigm change. The nature of associa-tions and the direction of causality in cases of “secondary insomnia” were un-certain. Furthermore, defining insomnia as “secondary” might lead to dimin-ishment of the importance of treating insomnia and may lead to only treatment of the “primary” condition with inadequate effect on the insomnia disorder.27 There are no quantitative criteria for the insomnia symptoms in the current nosology systems. There is, however, recommended minimum insomnia cri-teria for severity of sleep onset or maintenance of >31 minutes.28
Table 2. Diagnostic criteria for insomnia/insomnia disorder in DSM-5 and ICSD-3
DSM-5 - Persistent insomnia ICSD-3 - Chronic insomnia disorder
A) A predominant complaint of dissatisfac-tion with sleep quantity or quality, associated with one (or more) of the following symp-toms:
1. Difficulty initiating sleep 2. Difficulty maintaining sleep
3. Early-morning awakening with inability to return to sleep
B) The sleep disturbance causes clinically sig-nificant distress or impairment in social, occu-pational, educational, academic, behavioral or other important areas of functioning
C) The sleep difficulty occurs at least 3 nights per week
D) The sleep difficulty is present for at least 3 months
E) The sleep difficulty occurs despite ade-quate opportunity for sleep
F) The insomnia is not better explained by and does not occur during the course of another sleep-wake disorder
G) The insomnia is not attributable to the physiological effect of a substance.
H. Coexisting mental disorder and medical conditions do not adequately explain the pre-dominant complaint of insomnia
A) The patient reports one or more of the following:
1. Difficulty initiating sleep 2. Difficulty maintaining sleep 3. Waking up earlier than desired
4. Resistance to going to bed at appropriate time 5. Difficulty sleeping without intervention B) The patient reports one or more of the following related to nighttime sleep difficulty
1. Fatigue or malaise
2. Impaired attention, concentration or memory 3. Impaired performance (social, familiar,
occupational, or academic) 4. Mood disturbance or irritability 5. Daytime sleepiness
6. Behavioral problems (e.g., hyperactivity, Impulsivity, or aggression)
7. Reduced motivation, energy, or initiative 8. Proneness to judgment errors or physical
accidents
9. Concerns about or dissatisfaction with sleep C) The symptoms in (A) and (B) occur at least 3 times/week for at least 3 months
D) The reported sleep-wake complaints cannot be explained purely by inadequate opportunity or cir-cumstance for sleep.
E) Sleep-wake difficulty in not better explained by another sleep disordered (intoxication and acute
Epidemiology
Symptoms of insomnia are very common, with up to around 30% of the gen-eral population reporting symptoms sevgen-eral nights a week.29-32 The prevalence of the diagnosis of insomnia disorder is around 6 to 12%, depending on what definition was used in the studies and the study population.30, 33, 34 Insomnia is 1,4 to 2 times more common in women than men and increases with age.33, 35, 36 Longitudinal studies indicate that a majority of those with insomnia have persistent symptoms with approximately 70% reporting symptoms after a year and 46% after three years.32, 37
Pathophysiology
The exact mechanisms for the development and perpetuation of insomnia are still not established. However, in recent decades a string of different behav-ioral, cognitive, genetic and neurobiological models have been put forth.38, 39 Predisposing genetic factors have been found through Drosophila flies-stud-ies, whole-genome analysis and twin studflies-stud-ies, suggesting a multigene involve-ment in the developinvolve-ment of insomnia.40-42 The presence of precipitating fac-tors such as a stressor (i.e. life events) and moderating facfac-tors such as age, sex and comorbidities are more likely to lead to insomnia in persons with a genetic predisposition and with neurobiological abnormalities such as co-activation of both sleep-and wake promoting areas at the same time and reduced GABA in the cortex.39 Insomnia is considered a state of chronic hyperarousal with increased high-frequency EEG-activation, elevated heart rate, increased body temperature, and hypothalamic-pituitary hyperactivity, leading to the typical symptoms of insomnia.43 According to the three-factor behavioral model, acute insomnia becomes chronic through perpetuating factors such as the prac-tice of non-sleep activities in bed and staying in bed while awake.39, 44 Another behavioral model is the stimulus control model that suggests that insomnia is related to stimulus dyscontrol. The stimulus (the bed) is no longer strongly associated with sleep because through conditioning the bed is also paired with other activities such as work, watching TV, eating, reading, etc., thus, decreas-ing the possibility of sleep bedecreas-ing the body’s response to the bed.39, 45 In addi-tion to the behavioral processes, there are proposed cognitive processes such as a disproportional worry over sleep-related issues that perpetuate insom-nia.46 Levenson et al reviewed the above mentioned mechanism and proposed an integrative model of the pathophysiology of insomnia39 (Figure 3).
Sleep-related breathing disorders
Sleep-related breathing disorders encompass a wide range of disorders with distinct characteristics and prognosis. The focus here will be the obstructive sleep-related breathing disorders and snoring. Thus, there will not be any fur-ther discussion on central sleep apnea, sleep related hypoxemia disorders, or related hypoventilation, which are all included in the definition of sleep-related breathing disorders.
Definitions
In the ICSD-3, under the category of sleep-related breathing disorders, ob-structive sleep apnea syndromes (OSA) is its own sub-category, encompass-ing adult and pediatric OSA while, snorencompass-ing is found under the subcategory “isolated symptoms and normal variants,” together with catathrenia.20 In DSM-5 OSA is called obstructive sleep apnea hypopnea and is sorted under the category of breathing-related sleep disorders, while snoring is not men-tioned as its own entity.21 Another umbrella term that is commonly used is sleep disordered breathing (SDB), which includes a range of conditions linked by narrow airways or loss of normal respiration during sleep.
The diagnosis of OSA is made based on the occurrence of ≥ 15 obstructive respiratory events (obstructive apneas, hypopneas, or respiratory effort related arousals) per hour of sleep, as measured with polysomnography or respiratory polygraphy. In the presence of associated symptoms or associated comorbid-ities (e.g., excessive daytime sleepiness, fatigue, breathing interruptions, ha-bitual snoring, atrial fibrillation, hypertension, stroke or cognitive dysfunc-tion), the occurrence of ≥ 5 obstructive respiratory events/hour of sleep is suf-ficient for the diagnosis of OSA.20, 47 Obstructive sleep apnea-hypopnea
Predisposing factors
Precipitating & moderating factors Neurobiological changes
Cognitive and behavioral processes Hyperarousal
Insomnia
(OSA) and snoring are linked by a narrowing of the upper airway and in-creased upper airway resistance. Some believe that there is a continuum of disorders according to the degree of severity, ranging from snoring to upper airway resistance syndrome (UARS) to OSA.48 UARS was described almost three decades ago as a disorder that is distinct from OSA and characterized by daytime sleepiness and awakening due to increased respiratory effort without apneas or hypopneas.49 There is still some controversy regarding UARS as a separate disorder, moreover, in the latest version of the ICSD-3, the use of the term upper airway resistance syndrome is discouraged and instead included as a part of OSA.20 Although snoring is considered to be one of the cardinal symptoms of obstructive sleep apnea, the majority of snorers do not fulfill the diagnostic criteria for obstructive sleep apnea.50 According to ICSD-3, the di-agnosis of snoring is made if the following criteria are met: the affected indi-vidual or his or her bed partner report respiration-dependent, generally inspir-atory acoustic phenomena during sleep. There is no complaint of a sleep dis-order that could be causally attributed to the snoring, and the diagnostics yield no indication of the presence of another sleep-related respiratory disorder.20
Epidemiology
As the majority of those with self-reported snoring do not have obstructive sleep apnea, the prevalence of habitual snoring is much higher than the prev-alence of OSA.50, 51 Because there is no universal definition of snoring, the reported prevalence rates are wide, ranging from 7.6 to 60% in different stud-ies.52-55 For OSA, the definition is clearer, but it has changed over time, which makes it difficult to compare the prevalence numbers over time. In a 2015 review, Franklin and Lindberg concluded that there was a prevalence of OSA, as defined by AHI>15 and associated excessive daytime sleepiness, which was 3-18% in men and 1-17% in women.56
Both snoring and OSA are more common in males than in females.52, 56-58 While the highest prevalence of both snoring and OSA can be found in the middle-aged.52, 59 AHI increases even after the age of 60 but without corre-sponding daytime symptoms or snoring.60, 61
The most important risk factor for snoring and OSA is obesity.52, 53, 59, 61 Other risk factors that have been implicated are smoking and alcohol.52, 62-64
Pathophysiology
Snoring and OSA are both conditions of the upper airways. The sound of snor-ing is generated by a narrowsnor-ing of the upper airways that leads to turbulent airflow and vibrations in the pharynx. In OSA, there is a greater degree of narrowing of the upper airway or a total occlusion as compared to snoring without OSA. Both anatomical and functional abnormalities are thought to
contribute to the compromise of the upper airway. Anatomical factors of im-portance are the craniofacial structure, soft tissue structures such as the tongue size, increased size of the pharyngeal walls and increased fat deposition due to obesity.65-67 The pharyngeal dilator muscles work to oppose the collapse of the upper airways, whereas the negative pressure during inspiration promotes collapse. It is believed that the balance between the two opposing forces is lost in patients with OSA.68 There are also respiratory control mechanisms that are thought to be a part of the pathophysiology of OSA such as the arousal threshold.69
Diagnostic methods
There are different methods to assess both sleep and daytime sleepiness. There are the objective measures: polysomnography, respiratory polygraphy, actig-raphy, the multiple sleep latency test and the maintenance of wakefulness test. Then, there are subjective measure: sleep diary and different questionnaires.
Objective assessment of sleep and daytime symptoms
Polysomnography and respiratory polygraphy
Polysomnography (PSG) is the golden standard for objective sleep measure and diagnosis of OSA and other sleep disorders.47, 70 “The Manual for the Scoring of Sleep and Associated Events” by the American Academy of Sleep Medicine is the preferred guide on how to perform and interpret PSG. PSG is a multiparametric test in which signals are measured through electroenceph-alography (EEG), electrooculography, electromyography, cardiovascular sig-nals (ECG), pulse oximetry, measure of airflow, and respiratory effort.70 The data obtained are scored (either manually or automatically) and presented as a hypnogram of the sleep stages, chin/leg movements, and respiratory factors such as desaturations, apneas, hypopneas, and abdominal/thoracic respiratory effort.
Although polysomnography is invaluable for the diagnosis of some sleep dis-turbances and patient groups, for individuals with a high pretest probability of moderate to severe OSA, the simpler respiratory polygraphy may be used. 71 In this patient group, portable monitoring relates very well to PSG, with an agreement level of around 90%.72 For patients with significant cardiorespira-tory disease, respiracardiorespira-tory muscle weakness, hypoventilation, and hiscardiorespira-tory of stroke or severe insomnia, some clinical guidelines recommend PSG rather than polygraphy for sleep assessment.47 Polygraphy must, at a minimum, in-clude measurements of four of the following: airflow, respiratory effort (meas-ured through movement of thorax/abdomen), blood oxygenation, head-or
body position, pulse, EMG-activity in the jaw muscles, arterial tonometry, actigraphy, ECG, and/or sound.70
Actigraphy
Actigraphy is a non-invasive method of measuring sleep/wakefulness through the assessment of movement. Actigraphs are commonly placed on the wrist and can record for days or weeks if necessary. There is some evidence that the actigraph may be able to distinguish patients with a sleep disorder from nor-mal controls but there is no evidence that it can differentiate between different types of sleep disorders.73 Actigraphy is considered most useful in the diagno-sis of sleep disorders caused by disruption of the circadian rhythm and to as-sess treatment effects in subjects with insomnia.73
Multiple sleep latency test and maintenance of wakefulness test
The multiple sleep latency test (MSLT) and the maintenance of wakefulness test (MWT) are tests that try to objectively quantify excessive daytime sleep-iness. The MSLT is used to measure physiologic sleepiness and is performed in a sleep lab the day after an all-night polysomnography. The subject is given several opportunities to nap in a quiet and dark environment, with the poly-somnography recording the time it takes to fall asleep. Indications for MSLT include diagnosing narcolepsy and hypersomnias of central origin.74
The maintenance of wakefulness test (MWT), on the other hand, measures the ability to stay awake in a quiet, non-stimulation situation. It may be used to assess response to treatment of conditions with daytime sleepiness and is also used to evaluate individuals, such as pilots, who must remain awake for safety reasons.74 Neither the MSLT nor the MWT correlate well with subject measures of sleepiness such as the Epworth Sleepiness Scale74 (see below).
Subjective assessment of sleep and daytime sleepiness
The most frequently used method for assessment of sleep, sleep disturbances, and daytime sleepiness in epidemiological research is by means of various questionnaires. Sleep diaries are well-used and recommended as a part of di-agnosis of insomnia in clinical practice and may be used in studies.75
Sleep diary
A 7 to 14 day sleep diary is considered the tool of choice for subjective as-sessment of sleep and evaluation of treatment.76 Previously there was no standardized format, but as of 2012, a workgroup, consisting of experts in the field, developed the Consensus Sleep Diary. The main purpose of the stand-ardization is to allow for comparisons across different studies. The questions included in the Consensus Sleep Diary are 1) the time of getting into bed; 2)
time of trying to fall asleep; 3) time it took to fall asleep; 4) number of awak-enings; 5) duration of awakawak-enings; 6) time of final awakening; 7) final rise time, and 8) perceived sleep quality assessed on a Likert scale.77 A subsequent study has suggested quantitative criteria for the diary, in order to differentiate between insomnia and normal sleep.78 This has been validated against the In-somnia Severity Index and measurement with actigraph.79
Pittsburgh Sleep quality index (PSQI)
The PSQI evaluates sleep disturbances and sleep quality in the prior month using 19 questions in seven different categories: sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep med-ications, and daytime dysfunction.80 A combination of open-ended questions and Likert-type questions is used. The open-ended questions are converted into scores, and the sum of all questions yields a global score of 0 to 21. A total score ≥5 is considered an indicator of poor sleep quality and has a sensi-tivity of 89.6% for distinguishing between “good” and “poor” sleepers.80 The PSQI has a high test-retest validity and has also been validated in various pop-ulations.81, 82
Basic Nordic sleep questionnaire (BNSQ)
The BNSQ was developed by the Scandinavian Sleep Research Society in an attempt to standardize sleep questionnaires used in the Nordic countries.83 The BNSQ comprises of 21 questions, including questions on insomnia symptoms, sleep quality, use of sleeping pills, daytime sleepiness, tendency to fall asleep during the day, sleep time, bed time, wake-up time, naps, and snoring. The questions are answered using a five-point scale on how many nights/days per week something occurs. BNSQ has frequently been used in large-scale epide-miological studies.84-86
Insomnia severity index (ISI)
The ISI is used to assess various aspects of insomnia and to follow-up on treat-ment of insomnia.87 It comprises seven questions on insomnia symptoms, sleep dissatisfaction, daytime consequences, distress, and whether sleep prob-lems are noticeable to others. A 5-point Likert-scale is used, and a score of ≥10 has a high sensitivity and specificity for detecting insomnia.88 The ISI has been validated against sleep diary measures, and changes in ISI correlate with those in the sleep diary and polysomnography.87
Epworth sleepiness scale (ESS)
The ESS presents eight different activities and the respondent is asked to rate their risk of falling asleep while engaged in the activities.89 A 4-point scale is used, and a total tallied score of 11 or more represents “excessive daytime sleepiness”. The ESS has been widely used during the thirty years following
its development, yet a review of the measurement properties of the scale in 2014 showed a lack of high quality studies on its psychometric qualities.90
Consequences and co-morbidities
Sleep has many important functions, such as consolidation of memories and learning, energy conservation, as well as cardiovascular and immune system regulation, among others.91 It is therefore not surprising that the disruption of normal sleep may lead to negative health effects and be associated with comorbidities.
Daytime sleepiness
The definition of excessive daytime sleepiness (EDS), according to the ICSD-3, is “the inability to maintain wakefulness and alertness during the major waking episodes of the day, with sleep occurring unintentionally or at inap-propriate times almost daily for at least three months” .20 Other terms used in the context of EDS are hypersomnolence and hypersomnia. In the ICSD-3, hypersomnolence refers to the symptom of EDS, while some of the specific central disorders of hypersomnolence have hypersomnia in their names (e.g., idiopathic hypersomnia, hypersomnia due to a medical disorder).20 While EDS indicated high sleep propensity, fatigue “is a persistent sense of physical, emo-tional, and/or cognitive tiredness or exhaustion that can interfere with function and be disproportionate with prior activity”.92 There may be an overlap of both symptoms; furthermore, the terminology used by patients to describe their symptoms may make it difficult to distinguish between EDS and fatigue. In a quantitative study, subjects with OSA described the daytime consequences in various terms; feeling tired, having brain fog, feeling out of it, sleepy or fa-tigued.93 Even though both daytime sleepiness and fatigue are included in the diagnostic criteria for insomnia, in clinical practice patients with insomnia more often report fatigue and more rarely EDS.94
EDS is a common complaint in the general population with reports of a prev-alence of 8.2 to 33%.95-99 The wide span of the reported prevalence may be explained by different definitions of EDS and differences in study popula-tions. Some studies use ESS ≥ 10 to define EDS, while others used questions regarding EDS with a scale grading its severity and/or frequency.95-99 A sig-nificant number of people have persistent EDS over time. Fernandez-Mendoza et al, found that 38% had persistent EDS during a 7.5 year follow-up, and Jaussent et al. found a 33% persistence over 5 years of repeated annual assess-ments.99, 100 Even though EDS is a common complaint in primary care, it is probably underdiagnosed as all patients suffering from EDS do not seek med-ical advice. One study found that 45.5% of individuals with severe EDS and
30.7% of those with moderate EDS considered it to be a problem, but only 14% and 10.7%, respectively, had sought medical advice.97
Brown and Makker have an appealing and clinically relevant division of the possible underlying causes of EDS.101 They suggest that EDS can be a mani-festation of insufficient sleep duration, disturbed sleep, or disorders affecting the sleep wakefulness center in the brain.101 In addition, there are non-sleep related causes of EDS. Table 3 provides examples of all four suggested cate-gories of underlying causes.
Table 3. Examples of causes of excessive daytime sleepiness
Insufficient
sleep Disturbed sleep Sleep-wakefulness center Non-sleep related
Social or work pressure Environmental noise or light Neurological disor-ders Anemia (Insomnia) Medical condition (i.e.,
pain, asthma, GERD)
Side effects from medication
Electrolyte imbal-ance
Thyroid disease
Shift work OSA Narcolepsy Marijuana
Periodic limb movement syndrome
Idiopathic hyper-somnia
Prescription medi-cations
Alcohol Circadian rhythm
disorders
Alcohol
Insufficient amount of sleep is a common cause of EDS and can have different underlying causes. Somewhat surprisingly, long sleep duration is also associ-ated with EDS.99, 102 EDS in spite of long sleep duration may be a sign of depression or other sleep disturbances. EDS is a common complaint in shift-workers and is caused by both insufficient and decreased sleep quality.103 If not addressed, the sleep problems may lead to the development of chronic problems in the form of the circadian rhythm sleep disorder “shift work sleep disorder,” which is characterized by EDS and/or insomnia.20
Weight gain is associated with incident EDS, even in the non-obese.99, 102, 104 Even though obesity is strongly associated with OSA, it is also a significant predictor of EDS, independent of OSA.99, 105, 106 In most OSA patients, PAP-treatment eliminates EDS; however, a non-negligible number of patients have residual EDS despite of adherence to PAP-treatment.107
EDS is a common complaint in neurological disease such as epilepsy, Alz-heimer’s disease, Parkinson’s disease, myotonic dystrophies, and multiple sclerosis108. In Alzheimer’s disease, EDS may be caused by degenerative changes in the areas of the brain involved with sleep-wake regulation.108 In the other neurological diseases the etiology is believed to be multifactorial
involving comorbid sleep disturbances, sleep fragmentation due to the disease or symptoms secondary to the disease (such as pain and spasticity in MS).108 Depression, anxiety and chronic pain are predictors of EDS, depression is also associated with sleep disturbances such as insomnia and OSA.92, 97, 99, 100, 102, 109, 110
Alcohol consumption may interfere with nighttime sleep and cause next-day sleepiness but also act directly as a daytime sedative.111 Tricyclic antidepres-sants, hypnotics, and migraine medications are also associated with EDS.97 EDS is also a side effect of some antihypertensive medications such as beta-blockers and alpha agonist as well as anticonvulsive and antipsychotics med-ications.
In summary, EDS can be a part of, or secondary to, all kinds of sleep disorders as well as various non-sleep related problems and illnesses or a combination of different underlying causes. EDS can lead to increased accident risk (see section on accidents below). It is also a predictor for both fatal and non-fatal cardiovascular disease, of cognitive decline, and cardiovascular mortality in the elderly.112, 113 Furthermore, EDS is associated with reduced quality of life and a decreased desire to engage in social activities.114, 115
Accidents
A large proportion of motor vehicle accidents are attributed to sleep disorders, sleep deprivation, and driving while drowsy.116-119 Apart from the risk of inju-ries to the driver or bystanders, there are substantial societal costs associated with motor vehicle accidents caused by drowsiness.120 In addition to treatment of sleep disorders, there are suggestions of external adaptations such as con-tinuous monitoring of a driver through physiologic measures, vehicle adap-tions, giving warning for signs of unsafe driving, and road adaptaadap-tions, for example, with warnings of lane departure to decrease accident risk.121
Sleeping less than 6 hours per night is associated with an increased motor ve-hicle accident risk of 33% compared to those sleeping 7 to 8 hours a night.122 Among the sleep-disorders, OSA is the most examined regarding risk of motor vehicle accidents, even though other sleep disorders such as insomnia have also been implicated. Most studies have shown a substantially increased risk of motor vehicle accidents in those with OSA. A crash risk from 1.5 to almost 5-fold compared to non-OSA controls is reported, and treatment of OSA with CPAP decreases that risk.123-126 The Epworth sleepiness scale is widely used in clinical practice for the subjective evaluation of excessive sleepiness, espe-cially in OSA-patients. However, the scale does not correlate with the risk of motor vehicle accidents in OSA patients, which makes it hard to predict which
patients are at risk for accidents.123 Although data on insomnia and motor ve-hicle accident risk are scarce, one study showed that insomniacs had a 3-fold increased risk for having multiple accidents compared to matched good sleep-ers.127
While motor vehicle accidents and sleep have been the subject of many stud-ies, occupational accidents and their relation to sleep have been less explored. In a 2016 systematic review, Garbarino et al concluded that the odds of an occupational accident was nearly doubled in those with OSA. Eight of the ten studies in the review included only occupational traffic accidents. Of the two remaining studies, the study by Jurado-Gámez et. al did not show an increase of occupational accidents in workers with OSA, while the other one by Accat-toli et al did show an increased risk.128, 129The latter conclusion is supported by other studies, such as a study by Lindberg et al. which concluded that sleep-disordered breathing (defined as snorers with EDS), was associated with an increased risk of occupational accidents.130 Ulfberg showed that the risk of an occupational accident was increased both in heavy snorers and in those with verified OSA.131
Regarding occupational accidents and other sleep-disorders than OSA, there are several of cross-sectional and case-cohort studies conducted but very few prospective studies. A summary of the relevant studies can be found in Table 4. In cross-sectional, population-based studies, Daley et al132 found that in-somnia was associated with an increased likelihood of non-motor-vehicle ac-cidents, Kling et al133 found that those reporting poor sleep had an increased number of accidents and Shahly et al134 found an association between insom-nia and workplace accidents. Similar results were reported from other cross-sectional, population-based studies.135-137 In a review regarding the health ef-fects of shift work, Kecklund and Axelsson concluded that there is an associ-ation between shiftwork and occupassoci-ational accidents and that is mediated by both an insufficient amount of sleep and impaired sleep quality.103
An association between insomnia and sleep-problems and occupational acci-dents have also been found in specific populations in both cross-sectional and case-control studies. Examples of groups that have been studied are industry workers, patients with sleep disturbances in primary care, health plan sub-scribers, railway workers, construction workers, white collar employees, and veterinarians.138-147 In prospective studies, Åkerstedt et al found “that difficul-ties in sleeping during the last two weeks” was associated with fatal occupa-tional accidents.148 Also, Salminen et al. found that difficulty initiating sleep was associated with accidents in female workers in the public sector.149 Even though there are strong indications that insomnia is associated with occupa-tional accidents, there is a paucity of prospective population-based studies in the field.
Table 4 . Studies inve stiga ting occ upation al ac cide nts and s leep -r el ated problem s Study Study d es ign Study p opulation Definition of sleep d is tu rb an ce M ain findings Yamauchi 146 Cross-sectional Industr y workers (n=18,682) PS Q I ≥ 6 Sl eep-re lat ed pr oblem s ar e as so ci at ed w it h injur ies (OR 2.2) and n ear-m isse s (OR 1.7). Legér 150 C ros s-sectional Subjects wi th s le ep disturbances in p rimar y care (n=5,293) Insomnia accord ing to DSM-IV criteria Insomni a wa s assoc ia te d wi th a hi ghe r rat e of work acc idents . Sha hly 134 Cross-sectional Populations-bas ed (n=4,991) Insomnia accord ing to ICSD-2 and ICD-10 criteria Ins om nia w as a ss ociat ed w ith w orkplac e ac ciden ts (OR.1.4). Ke ssle r 147 C ros s-sectional Health pl an subs cribers (n=4,991) Insomnia accord ing to ICSD-2 and ICD-10 . Insom nia w as associat ed w ith oc cupat ional injuri es (OR 1.9). Kling 133 Cross-sectional Population-bas ed (n=69,584) Having DIS or DMS so m e-tim es or m ost of the tim e There was an a ss ociation between trouble sleepin g m ost of the tim e and oc cupa tio nal injur y in bo th men and women (OR 1.6 men, 1.5 w omen) and fo r
women trouble sleep
ing sometimes (OR 1.3)
. Salm inen 149 Prospective coh ort Public se ctor workers (n=40,386) H aving D IS , D M S, EM A or non-refreshing sleep 5-7 nights a week Non-refreshing sleep i s a pr edictor of occupation al injuries in men ( O R 1.5) bu t not women. DIS is a predic to r f or occup ation al injur y in wom en (OR 1.6) bu t not men. Daley 132 Cross-sectional Population-bas ed (n=953) Insomnia d ef ined as dissa tis-faction with sleep and h aving D IS , D M S, or E M A ≥ 3 nights/week ≥1 month and da ytim e im pair m ent Ins om nia w as a ss ociated w ith an in cre as ed lik el
i-hood of non-motor-vehicle accid
ents (OR 2
nson 136 Cross-sectional Population-bas ed (n=1,000) H aving D IS , D M S, or EM A at leas t a few n ig hts /week and EDS at leas t a few day s/week In so m ni a s ym pt om s w er e a ss oc ia te d w it h o cc up
a-tional accidents during th
e p as t y ear (OR 2.28) . 144 C ross-sectional W orkers in small and m e-dium-scale enterprises, 92% in industrial sector (n=2,903) DIS, DMS or EMA ≥3 times/week. Sleeping not so we ll o r ve ry p oo rl y at n ig ht .
Somewhat or definite insuffi- cien
cy in s le ep. DIS (OR 1.5), sleeping poorly a t night (OR 1.5) and sleep insufficien cy (OR 1 .3) or a ny insomnia sy m p-tom (OR 1.5) w ere associated w ith higher prev a-lence of occupati onal accid ent. 143 C as e-contro l Railwa y work ers (n=2,610) ≤6 hours of sleep per day , and/or not sleeping well and/or regular consumption of h ypnotics) Sleep d is orders w ere as so cia ted w ith having ≥ 2 oc-cupational injur ies (OR 1.99) but not w ith a single injur y. 135 C ross-sectional Population-bas ed (n=7,980) Sl eep prob lem s in t he l as t 14 day s Minor injur ies at w ork in the last ye ar w ere associ-ated with report ing s leep problems (OR 1.61). 145 Cross-sectional Manufactur ing workers (n=1,298) DIS, DMS or EMA ≥ 3 tim es/week Insomnia sy m ptoms were correlated with occup a-tional in jur y (O R 1.64) . 141 C as e-contro l Railwa y work ers (n=2,610) ≤6 hours of sleep per day , and/or not sleeping well and/or regular consumption of h ypnotics. Sleep d is orders w ere as so cia ted w ith ac ciden ts r e-lated to ph ysical exertion and pa in due to movement (OR 2.23). 142 Case-contro l Construction workers (n=1,760) ≤6 hours of sleep per day , and/or not sleeping well and/or regular consumption of h ypnotics. Sl eep dis orders w ere as so ciat ed w ith oc cupa tion al acc idents ( OR 1. 92).
Doi 140 C ros s-sectional W hite co ll ar em plo yees (n=4,868) PS Q I ≥ 5 Poor s leepers h ad an in cre as ed r is k of occup ation al acc idents ( OR 1. 48). Åkers tedt 148 Prospective coh ort Population-bas ed (n=47,860) Affirm ative ans w er to “ have you had diffi cul ties in sl eep-ing during th e last two weeks ” Sleep dis turban ce is a p red icto r of fa ta l occup a-tional accident s (RR=1.76). Léger 137 C ross-sectional Population-bas ed (n=1,598) A t l ea st 2 o f D IS , D M S, EMA or non-r efr eshing sleep, ≥3 tim es/week for at le ast 1 month The insomnia group re porte d 7 time s more oc cu pa
-tional accidents than t
he non-insomnia group. Gabel 139 C as e-contro l Veterin arians (cases=193) ≤ 6 hours of sleep Short sleep le ng th w as a ssociated w ith a n in cr eas ed risk of occupatio nal injur y (RR=1.8). Lavie 138 Cross-sectional Industr y workers (n=1,502) DIS, DMS, DIS+DMS, E DS, DIS+DMS+EDS many times or alway s Ins om nia sy m ptom s are as so ciated w it h an in-creas ed num ber of oc cupat ional acc idents ; th e as -so ciation is g reat er w hen ins om nia sy m ptom s are in com bination wit h EDS. PSQI=Pittsburg sleep quali ty in dex, OR=odds r adio, DSM = Di agnostic and sta tistica l manual of men tal disord ers, ICSD=Inter natio nal classification of sleep disorders, ICD-10=Intern ational class ification of diseases, DIS=dif ficu lty in itia ting sleep, DMS=diff icul ty mainta ining s le ep, EM A=early morning awakening , EDS=exces si ve da yt im e s leep ines s.
Gastroesophageal reflux and sleep
Gastroesophageal reflux disease (GERD) is defined as a condition that devel-ops because of reflux of the gastric content into the esophagus causing typical symptoms.151 The most common symptoms of GERD are heartburn and re-gurgitation.151 An estimated prevalence of at least weekly GERD symptoms in the general population is approximately 13%, though there are geographic variations.152 The highest prevalence is measured in North-America, 19.8%, while in Europe, the corresponding estimated prevalence is 15.2%.152 Risk factors for GERD include obesity, certain types of exercise (e.g., stooped po-sition, weight lifting), diet, smoking, and asthma.153Age and gender are not established risk factors for GERD, but they are associated with an increased risk of GERD complications.153
In most cases, the diagnosis of GERD is made clinically without invasive ex-aminations. In the presence of typical GERD symptoms, without alarm symp-toms (dysphagia, odynophagia, weight loss, GI-bleeding, or anemia) an upper endoscopy of the GI-tract is not required for diagnosis of GERD.154 Guidelines for gastroesophageal reflux recommend establishing a diagnosis on the basis of patients’ self-reported symptoms.154 In older patients, and in patients with alarm symptoms, endoscopy is useful and might reveal the presence of com-plications.154 Diagnostic questionnaires have been developed for GERD, but a comparative study of endoscopy, pH-measuring, and PPI-trial to define GERD showed a better sensitivity and specificity for physician-diagnosed dis-ease compared to questionnaire diagnosed disdis-ease.155
Gastroesophageal reflux during nighttime has been suggested to be of specific interest. Physiological changes during sleep, such as a decline in saliva pro-duction, decreased frequency in swallowing, gastric emptying, and peristalsis lead to the diminished ability to prevent and neutralize reflux events.156 The majority of GERD patients have both daytime and nighttime symptoms. Noc-turnal gastroesophageal reflux (nGER), with symptoms such as trouble getting to sleep because of reflux symptoms, awakening at night because of reflux symptoms or experiencing heartburn symptoms when awakening in the morn-ing, is reported by 65% to almost 90% of GERD patients.157-160 Those with nGER have more sleep disturbances and a greater degree of daytime impair-ment than those with only daytime symptoms.159-162 In addition, nGER is as-sociated with a more severe GERD and with increased likelihood of compli-cations such as erosive esophagitis, stricture, ulcer, Barret’s esophagus and adenocarcinoma of the esophagus.163-165
Numerous studies have found a significant correlation between sleep disturb-ances and GERD. In a prospective randomized study, Yamasaki and
col-leagues found that two nights of sleep deprivation induced a significant in-crease in acid exposure time, both among healthy individuals and individuals with GERD.166 Matsuki found that patients with GERD were more often sleep deprived than healthy controls.167 Increasing duration and frequency of reflux events are correlated with increased symptoms of difficulties initiating and maintaining sleep.168 Janson et al found a more than 3-fold risk of GERD among insomniacs compared to normal sleepers.169 Patients with nGER and sleep complaints had a greater duration of acid exposure in the esophagus, and sleep deficiency can lead to hyperalgesia in the esophagus, thus worsening heartburn symptoms.170, 171 Lindam et al. found a strong increased risk of new-onset GERD among those with insomnia and a moderately increased risk of new-onset insomnia among those with GERD, suggesting a bidirectional as-sociation between the two conditions.172 Furthermore, benzodiazepines widely used for the treatment of insomnia may lead to an increase in the acid contact time during reflux events through slowing of gastric emptying and diminished arousal response due to its sedative effect.173 Further evidence of the portance of nGER on sleep is that several studies have demonstrated an im-provement of sleep quality, sleep efficiency, and fewer arousals and awaken-ings with treatment of nGER with proton pump inhibitors.174, 175
The association between sleep disturbances and GERD is especially promi-nent among individuals with SDB, as described by a recent meta-analysis.176 Patients with OSA have more reflux events than controls, but less than half of the apneas and arousals are related to reflux events.177 One explanation for the lack of a direct correlation between apneas and reflux events could be that the repetitive strain associated with increased respiratory effort in OSA may weaken the gastroesophageal junction, causing a chronic predisposition for reflux events.178 Moreover, in turn, nGER causes upper airway edema that ex-acerbates OSA.179 The use of CPAP in OSA-patients leads to an improvement of reflux but CPAP also reduced the reflux parameters in healthy controls in-dicating a non-specific treatment effect.177, 180 Whether there is a causal link-age between GERD and SDB or if they merely share common risk factors such as obesity is still up for debate. Emilsson et al found that after adjusting for BMI, snoring was associated with nGER but not AHI was.181
The correlation between night time reflux events and arousals and awakenings differs between studies. Also, sleep maintenance problems can be a part of insomnia symptomology or a consequence of OSA. Dickman et al. found that the vast majority of acid reflux events during nighttime preceded short arous-als,168 while other studies suggest that the majority of reflux events follow arousals or conscious awakening periods, suggesting that reflux was not the primary cause of the events; rather it was secondary to arousals/awakenings from other causes.182, 183 Repeated awakenings during the night may lead to daytime sleepiness, and GERD, especially nGER, is associated with increased
daytime sleepiness.184 It is not well-established if the combination of SDB and nGER increases the risk of daytime sleepiness or if they are merely two con-ditions associated with the same life-style factors.
Cardiovascular diseases and sleep
During normal sleep in healthy individuals, several physiological changes in the cardiovascular system occur. At the beginning of the night, when NREM is initiated, there is an increase in parasympathetic activity and a decrease in sympathetic activity, resulting in a characteristic “dipping” of the blood pres-sure and reduction in the heartrate.185 During REM-sleep, blood pressure re-turns to levels comparable to wakefulness, but there is an instability and fluc-tuation between the parasympathetic and sympathetic parts of the autonomic nervous system with surges in blood pressure.185 Arousals, whether occurring in NREM or REM-sleep lead to temporary increases in blood pressure and heart rate.186 Since the majority of each nightly sleep period consists of NREM-sleep, the cardiovascular system, on average, is less active during sleep. It is therefore only natural that various forms of disrupted sleep may have cardiovascular consequences.
Insomnia is associated with an increased risk of ischemic stroke, myocardial infarction, and hypertension.142, 187, 188 Patients with chronic insomnia have non-dipping blood pressure and elevated nighttime blood pressure compared to good sleepers, which may be associated with increased cardiovascular risk.189-191 It is believed that the hyperarousal associated with insomnia causes increased sympathetic activity, decreased parasympathetic activity, increased cortisol levels, increased norepinephrine and interlekin-6 levels, which are thought to play a role in the association between insomnia and cardiovascular disease.43, 192 Another factor of importance may be a dysregulation of the hor-mones regulating hunger that may lead to weight gain and subsequent meta-bolic syndrome.193
Sleep-disordered breathing is associated with coronary heart disease and with the risk of developing stroke, hypertension, and fatal cardiovascular events. 194-196 Furthermore, snoring is independently associated with carotid atheroscle-rosis and increased all-cause mortality.197, 198 The underlying mechanism be-tween SDB and cardiovascular disease is not established, but the intermittent hypoxia associated with SDB is believed to cause oxidative stress and inflam-mation, which may contribute to the development of atherosclerosis.199, 200 In-termittent hypoxia and arousals are also associated with increased sympathetic activity, activation of the renin-angiotensin-aldosterone system, and the hypo-thalamus-pituitary-adrenal axis and thus contribute to the development of hy-pertension.201
There is a significant overlap between insomnia and SDB. Insomnia symp-toms are reported by up to 60% of patients with sleep apnea, and 29–67% of insomnia patients have OSA.202, 203 Only a few studies have examined the combination of insomnia and sleep apnea and risk of cardiovascular disease91, 204-206 (Table 5). Björnsdóttir et al. studied OSA-patients referred for treatment with CPAP and did not find an association between hypertension or diabetes and self-reported insomnia symptoms.206 In a population-based cross-sec-tional study, Vozoris et al. did not find any differences in risk of hypertension, diabetes, congestive heart failure, myocardial infarction, or stroke between those with sleep apnea combined with insomnia compared to those with sleep apnea alone.204 In contrast, Gupta et al, used a large data-base with data on patient visits and found that the group with visits for both insomnia and OSA were more likely to have hypertension and cerebrovascular disease.205 Simi-larly, there was a higher prevalence of cardiovascular comorbidity in OSA-patients with insomnia than those without in a study by Saaresranta et al.91 Thus, there is conflicting data regarding the effect of the combination of in-somnia and sleep-disordered breathing on the risk for cardiovascular dis-ease.91, 204-206 More studies are needed to shed light on if there is an association, and if so, how great.
e 5 . St udies regard ing OS A and insomnia an d card iovascu lar outcomes. Study d es ign Study p opulation Insom nia Obstructive sleep apnea Cardiovascular disease 206 C ross-se ctiona l OSA-patients ref erred for CPAP-treatment (n=824) & c ont ro ls ( n= 76 2) Self-reported d if ficulties initiating or maintaining sleep ≥ 3 tim es /week OSA (criteria no t sp ecifi ed) Patient-reported doctor diagnosed h yp er tension and diabe tes or us e o f m edica tion for h ypertension or diabetes 204 Cross-sectional Population-bas ed (n=12,593) Self-reported DI S or DMS
and/or EMA ≥ 5 tim
es/m onth. Self-reported h ealth professional-diagnosed slee p a pnea . Se lf -r ep or te d hea lth p ro fe ssi onal - diagnosed h yp er tension, diabe tes , m yoc ar dial infar ct ion, congestiv e heart failur e and
stroke. Measurement of blood pres
su re x 3 . 205 C ase-contro l Patients visiting for sleep apnea; data fro m national datab ase (n = 7,2 34 visits) Patient visi t with reason to visit state d as insomnia or disorder of initiating or m aintaining sle ep . Patient visits co ded for sl eep apne a. Codes for p ati en t visits; hy per tens ion , h ea rt f ailur e, is chem ic h ear t d is eas e, c ardi ac dy sr hy thm ia and cer ebrovas -cular dis eas e. rant a 91 C ros s-sectional Sl eep cente r pat ients w ith AHI≥5 (n=6,555 ) Self-reported DI S, sleep duration ≤6 h, use of h yp -notics or ph ys ician diag-nosed insomnia AHI≥5 Patient-reported cardiov ascular comorbidity (h ypertension , is chem ic h ear t d is eas e and cerebrov as cul ar dis eas e). e sle ep apn ea, C PAP=con tinuou s positiv e airway pre ssure, AHI= apnea hypopn ea inde x, DIS=diff ic ult y in itia ting sl ee p, DMS=diff ic ulty slee p, EM A=early m orning awaken in g.
Cigarette smoking and sleep
Cigarette smoke contains around 5000 different compounds whereof there are many that have negative cardiovascular and respiratory effects or are carcino-genic.207 Nicotine is considered the primary pharmacologically active compo-nent of cigarette smoke and acts on the central nervous system by the activa-tion of nicotine-acetylcholine receptors.208 The activation of the receptors leads to the release of several neurotransmitters that are involved in the regu-lation of sleep-and-wake (dopamine, serotonin, acetylcholine, norepineph-rine, GABA), which disrupt the normal equilibrium that exists and may lead to insomnia-like symptoms.208 Due to the short half-life of nicotine, another mechanism for occurrence of some of the sleep related problems may be caused by acute withdrawal.209
Polysomnographic studies have shown that smokers require more time to reach sleep stages I and II (i.e., difficulty initiating asleep) and lower sleep efficiency compared to non-smokers.210, 211 Smokers also reported more sleep problems with standardized sleep quality questionnaires.210, 212-215 They had more complaints of sleep-latency and a shorter reported sleep duration.210, 212-215 There are also increased reports of difficulty maintaining sleep and early morning awakening among smokers.214 Smoking during nighttime has been associated with more insomnia symptoms and a shorter sleep duration com-pared to smokers who do not smoke during the night.215 In addition to insom-nia symptoms, smoking is also associated with sleep-related movement disor-ders and obstructive sleep apnea.210, 216, 217
Different factors impacting the success of smoking cessation have been pro-posed and studies show that positive predictors may be lower nicotine depend-ence, higher age, higher education and fewer years of smoking.218-220 Sleep disturbances, as a part of nicotine-withdrawal, have been indicated as predic-tors of short term-relapse.218, 219, 221 However, it is not established how preex-isting sleep-disorders in smokers affect the chances of successful smoking cessation, knowledge that would be of interest when developing programs supporting smoking cessation.
Aims
The overall aim of this thesis was to investigate the impact of sleep-related complaints (insomnia symptoms and snoring) on various consequences and comorbidities.
I To assess the association between insomnia symptoms and self-reported and register-reported occupational accidents in women during a 10-year follow-up period.
II To compare the occurrence of daytime sleepiness and
involuntarily falling asleep in women with and without the combination of nGER and snoring at baseline/follow-up. III To analyze if sleep disturbances are a predictor of smoking
ces-sation and if continued smoking associates with the develop-ment of sleep disturbances at follow-up.
IV To investigate the association of a combination of insomnia symptoms and snoring with comorbidity, daytime sleepiness, and use of hypnotics.
Methods
Study populations
SHE (Papers I & II)
Sleep and Health in Women (SHE) is a population-based prospective study that was initiated in the year 2000. A baseline postal questionnaire was sent to 10,000 randomly selected women aged ≥ 20 years in the Municipality of Upp-sala, Sweden. The questionnaire was completed by 7,051 women, yielding a response rate of 71.6%.222 Ten years later, a follow-up questionnaire was sent to the 6,455 participants who were still living in Sweden. The follow-up ques-tionnaire was submitted by 5,193 women, yielding a response rate or 80.5%.110 The study design is illustrated in Figure 4.
In paper I, the study cohort comprised 4,320 women who answered both ques-tionnaires, and in the follow-up questionnaire they reported working ≥10 hours/week for ≥1 year during the time between the questionnaires. Women
Baseline stu
dy
2000
Follow-up study
2010
Participants in baseline study (n=7,051)
Follow-up questionnaire (n=6,455) Excluded (n=596) Baseline questionnaire (10,000 women aged ≥ 20)
Response rate 71.6%
Deceased or moved to unknown address.
Response rate 80.5%
Deceased or moved to unknown address.
Questionnaire sent by mail
Excluded (n=152)
Participants in follow-up study (n=5,193)
